# Robust, Deep and Inductive Anomaly Detection

arXiv: https://arxiv.org/abs/1704.06743 · HTML: https://mlsanity.com/paper/1704.06743/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Apr 2017.

## Built on by (most cited first, 24 of 24)

- [Learning Normal Dynamics in Videos with Meta Prototype Network](https://mlsanity.com/paper/2104.06689/) (2021)
- [Rethinking Reconstruction Autoencoder-Based Out-of-Distribution Detection](https://mlsanity.com/paper/2203.02194/) (2022)
- [A Survey on Open Set Recognition](https://mlsanity.com/paper/2109.00893/) (2021)
- [Utilizing XAI technique to improve autoencoder based model for computer network anomaly detection with shapley additive explanation(SHAP)](https://mlsanity.com/paper/2112.08442/) (2021)
- [Multi-scale Cross-restoration Framework for Electrocardiogram Anomaly Detection](https://mlsanity.com/paper/2308.01639/) (2023)
- [Enhancing Unsupervised Anomaly Detection with Score-Guided Network](https://mlsanity.com/paper/2109.04684/) (2021)
- [Can self-training identify suspicious ugly duckling lesions?](https://mlsanity.com/paper/2105.07116/) (2021)
- [Vision-Language Models Assisted Unsupervised Video Anomaly Detection](https://mlsanity.com/paper/2409.14109/) (2024)
- [Adversarial Support Alignment](https://mlsanity.com/paper/2203.08908/) (2022)
- [Latent-Insensitive autoencoders for Anomaly Detection](https://mlsanity.com/paper/2110.13101/) (2021)
- [Self-Supervised Training with Autoencoders for Visual Anomaly Detection](https://mlsanity.com/paper/2206.11723/) (2022)
- [What is Wrong with One-Class Anomaly Detection?](https://mlsanity.com/paper/2104.09793/) (2021)
- [EARLIN: Early Out-of-Distribution Detection for Resource-efficient Collaborative Inference](https://mlsanity.com/paper/2106.13842/) (2021)
- [Outlier Detection using AI: A Survey](https://mlsanity.com/paper/2112.00588/) (2021)
- [Spatio-Temporal Relation Learning for Video Anomaly Detection](https://mlsanity.com/paper/2209.13116/) (2022)
- [Spot The Odd One Out: Regularized Complete Cycle Consistent Anomaly Detector GAN](https://mlsanity.com/paper/2304.07769/) (2023)
- [Building One-class Detector for Anything: Open-vocabulary Zero-shot OOD Detection Using Text-image Models](https://mlsanity.com/paper/2305.17207/) (2023)
- [Anomaly Detection in Electrocardiograms: Advancing Clinical Diagnosis Through Self-Supervised Learning](https://mlsanity.com/paper/2404.04935/) (2024)
- [Normal Learning in Videos with Attention Prototype Network](https://mlsanity.com/paper/2108.11055/) (2021)
- [Diminishing Empirical Risk Minimization for Unsupervised Anomaly Detection](https://mlsanity.com/paper/2205.14676/) (2022)
- [Robust Autoencoders for Collective Corruption Removal](https://mlsanity.com/paper/2303.02828/) (2023)
- [Technical outlier detection via convolutional variational autoencoder for the ADMANI breast mammogram dataset](https://mlsanity.com/paper/2305.12068/) (2023)
- [Rethinking Unsupervised Outlier Detection via Multiple Thresholding](https://mlsanity.com/paper/2407.05382/) (2024)
- [Noise & pattern: identity-anchored Tikhonov regularization for robust structural anomaly detection](https://mlsanity.com/paper/2511.07233/) (2025)

## Abstract (arXiv)

PCA is a classical statistical technique whose simplicity and maturity has seen it find widespread use as an anomaly detection technique. However, it is limited in this regard by being sensitive to gross perturbations of the input, and by seeking a linear subspace that captures normal behaviour. The first issue has been dealt with by robust PCA, a variant of PCA that explicitly allows for some data points to be arbitrarily corrupted, however, this does not resolve the second issue, and indeed introduces the new issue that one can no longer inductively find anomalies on a test set. This paper addresses both issues in a single model, the robust autoencoder. This method learns a nonlinear subspace that captures the majority of data points, while allowing for some data to have arbitrary corruption. The model is simple to train and leverages recent advances in the optimisation of deep neural networks. Experiments on a range of real-world datasets highlight the model's effectiveness.